Cric.ac.uk

In this study, we examine the distribution of revenues for a comprehensive sample of new drugs introduced into the United States during the period, 1988 to 1992. In earlier research, we examined the returns to R&D for the U.S. new drug introductions during the 1970s and early "The Distribution of Sales from Pharmaceutical Innovation"
1980s.[1,2] One of the key findings was that the top decile of new drugs accounted for a large share of the total market value generated by these entities. In this regard, the returns to R&D projects in pharmaceuticals have similar properties to that of venture capital investments. This has important implications for both private and public decision makers.
A new analysis of this issue is warranted by a number of important changes that have been occurring on both the demand and supply sides of the market for new drugs. In particular, there has been significant new entry and industry restructuring since our last analysis of the returns to R&D. In addition, managed care has grown dramatically during the 1990s and now accounts for a dominant share of drug prescriptions. These factors can significantly affect sales life cycles and the distribution of revenues across new drug introductions.
preliminary version of this paper was presented at a conference in Talloires, France in July 1999 that was hosted by the Tufts University Center for the Study of Drug Development.
Do not quote or reference without permission from the authors.
B. Returns for Venture Capital Investments and IPOs
I. Background
F. M. Scherer has shown in recent work that many other innovational activities are A. The 1980-84 Cohort of New Drug Introductions
characterized by skewed outcome distributions.[3,4] Of particular interest are two of his data In this section, we summarize some of the core findings from our prior work on samples that involve a large number of investments by U.S. venture capital firms in start-up pharmaceuticals and relate it to recent work on the returns for venture capital investment. Our last companies between 1969 and 1988. The first data sample was compiled by Venture Economics analysis focused on a comprehensive sample of 64 new chemical entities (NCEs) introduced into Incorporated and involved a portfolio of investments in 383 start-up companies made by 13 venture the U.S. market between 1980 and 1984. In this regard, Figure 1 shows the sales profiles over the capital firms. The second sample involved a similar data set assembled by Horsley-Keough marketing life cycle for the top two deciles of NCEs (ranked by 10th-year U.S. sales) and the mean Associates of 670 distinct investments made by 16 venture capital companies.
and median compound. This diagram indicates that there is a high degree of variability in the sales Scherer's analysis indicates that investment returns from venture-financed start-ups are performance of NCEs. In particular, the peak U.S. sales for the top decile compounds exceeded highly skewed. As shown in Table 1, a relatively small number of start-up firms generate a large 700 million dollars, the second decile was approximately 300 million dollars, the mean compound share of the total investment value, as measured by the capital appreciation or loss at the time of 150 million dollars and the median compound only 50 million dollars. investor exit from each investment. In the case of the Venture Economics sample, 62% of the total also estimated the "quasi profits" for each entity--the surplus of global sales revenues over value generated by all 383 investments was accounted for by the most profitable decile of projects. production and distribution costs and discounted them to the date of market launch. The top decile- For the Horsley-Keough sample, 59% of the overall value was attributed to the top decile of start- -the most profitable ten percent of the compounds--contributed 48% of the quasi profits realized by up company investments. This may be compared to our samples of 1980 NCEs where 48% of the the full sample of NCE introductions in this period. By contrast, the bottom half of the distribution quasi profits were accounted for by the top decile of NCEs.
(deciles 6 through 10 encompassing the entities with peak sales below 50 million dollars), Scherer also examined the stock market performance of a comprehensive sample of 131 accounted in total for only 8 percent of the quasi-rents.
venture-funded high-technology companies which had their initial public offerings between 1983 and 1986. He examined the returns a decade later from an equal dollar investment in each of these companies at the time of their IPO. An investment in a full bundle of these IPO companies would have slightly outperformed a comparable dollar investment in the NASDAQ index over the same period.1 However, the market performance of these IPO firms also exhibited the same tendency particular disease.[5] The pharmaceutical industry has also been characterized historically by toward extreme values as in the venture-financed start-up investments samples discussed above. significant first-mover advantages.[6,7] Other things equal, later market entrants tend to capture As shown in Table 1, the 14 firms that comprised the most profitable decile of these IPO companies accounted for 62 percent of the overall market value in 1995. Correspondingly, the other 123 high- The most novel compounds ex-ante face the greatest risks from a scientific, regulatory and tech firms in this sample accounted for the remaining 38 percent.
commercial perspective. In this regard, the therapeutic profile of these compounds are the most difficult to predict based on pre-clinical screens and leads. In addition, the long time lags and R&D C. Implications for R&D Investments
activities of competitors magnify these scientific and technical risks. Accordingly, unforeseen The data shown in Table 1 indicates that R&D investments in pharmaceuticals shares much clinical outcomes, the introduction of rival products and other changes in the market, and regulatory in common with private investments by VC firms in start-up companies as well as public market problems and lags can dramatically affect a new drug's economic prospects during the development investments in high-technology IPO companies. All of these innovational investment activities are characterized by a high degree of riskiness. This results from the fact that a few extreme values These factors help to explain why so many of the compounds in Figure 1 are marketed account for a large share of the cumulative realized returns. As Scherer and others have observed, despite very small peak sales revenues and with quasi-profits that are a small fraction of mean R&D the law of large numbers doesn't work very well when the probability distribution of outcomes is costs.2 If significant uncertainties surrounding a compound's economic prospects are not resolved highly skewed. One important consequence for pharmaceutical R&D is that considerable until clinical development is largely complete, most of the R&D costs are then sunk. At this point, variability in portfolio outcomes can be expected, even for those pharmaceutical companies with as long as a compound's expected revenues cover the incremental or variable costs on a prospective large diversified portfolios of R&D pipeline drugs.
basis, it will be rational to market or license out the compound, even if this doesn't cover any of the In the case of pharmaceuticals, the blockbuster compounds which comprise the top decile of NCEs in Figure 1 generally represent significant therapeutic advances in treating a particular and failure to meet NASDAQ financial criteria.
disease, usually one with significant market size. In most instances, these therapies are the first or 2We did not have R&D costs on an individual NCE basis. Another factor of course could be thatR&D costs are also lower for drug entities with smaller sales and quasi-profits. While this may be second introductions in a new chemical class of compounds and offer a novel approach to treating a the case, an analysis of R&D costs for a representative sample of NCEs at different stages of theR&D process by DiMasi et al. indicated that there is much less variability in R&D costs than in revenues across NCEs [8]. This is plausible, given the fact that all FDA approved drugs must pass 1Returns were based on the market values of these companies approximately one decade later stringent regulatory hurdles. Approved drugs also share in common pre-project discovery costs and (December 31, 1995). This analysis takes account of the market values of the surviving IPO the costs of failures. These components account for over 50 percent of the mean estimated R&D companies, those that merged with other firms, and those that were deleted because of bankruptcies compound's large fixed R&D costs. Of course, over the long run, the firm also must have its share managed care institutions.[9,10] PBMs have implemented drug formularies to encourage more price of winners for its R&D program to be profitable and remain viable.
competition and incentive programs for generic drug usage when brand products come off patent. At the same time, managed care institutions have broadened insurance coverage for prescription II. Recent Market Developments
drugs, and unit sales have grown as drug therapies and compliance have been encouraged as a way The basic sample to be investigated is 110 new drug entities developed for the U.S. market, of avoiding more expensive medical treatments.
approved by the FDA and introduced into the United States market between 1988 and 1992. This PBMs and HMOs can affect the sales revenues of a new drug introduction in alternative is a comprehensive sample of the new drug entities introduced into the United States market during ways over the marketing life cycle. New drugs that represent novel therapeutic interventions to this period. In this paper, we will be focusing on the U.S. sales performance of these entities. In particular diseases and conditions have generally received broad coverage and speedy approvals future papers, we will examine the returns on R&D to these entities and will integrate global sales onto drug formularies. However, as follow-on drugs are introduced into the same class, price discounting and competition usually occurs to obtain formulary access. The growth of managed In our past work, we have found that differences in sales revenues is the major driving force care has also been an important factor contributing to more rapid sales erosion when drugs come off underlying the skewed distribution of quasi-profits across NCEs. An analysis of sales performance patent. Therefore, as a new drug proceeds through its marketing life cycle, and as competition in the United States is therefore an interesting analysis in its own right. In this regard, the United develops in a given therapeutic class, the impact of PBMs of managed care providers on sales States is also the largest market for pharmaceuticals, accounting for roughly half of the sales of the revenues is subject to important shifts over time.
new drug introduction studied in past samples. We also found that the sales of these new drugs in other major markets (Europe and Japan) were significantly positively correlated with their U.S.
B. Biopharmaceuticals, Orphan Drugs and Supply Side Changes
There have also been important changes on the supply side of the market. In this regard, the number of new drug entities introduced into the U.S. market in the 1988-92 period is significantly A. Managed Care and Demand Side Changes
larger than for the early 1980-84 period. This reflects some important industry developments. As noted in the introduction, the demand side of the market for new pharmaceuticals has First, the current sample includes new biopharmaceutical entities as well as new chemical entities. been undergoing substantial change over the past decade. Pharmacy benefit management firms The biotech industry was essentially at an infant stage in the early 1980s. However, by the early (PBMs) have emerged as the main overseers of the prescription drug plans of employers and 1990s, it had become a significant source of new therapeutic entities.
Another important event was the passage of the Orphan Drug Act by Congress in 1983. III. Data Samples and Methodology
This provided incentives in the form of tax credits, market exclusivity, and regulatory assistance for Annual drugstore and hospital sales in the United States were obtained from IMS America the development of drugs targeted to diseases and conditions involving small patient for each of the 110 new drug entities in our sample. The sales data covered the period 1988 to populations.[11] In particular, a drug was eligible for orphan drug status under the law if it is 1998. This provided between 7 to 11 years of sales data for the drugs in our sample cohort, approved for an indication involving a population less than 200,000 patients. Roughly one-quarter depending on a drug's year of introduction.
of the drugs in our current sample were granted orphan drug status for at least one approved Twenty years was chosen as the expected market life for this cohort. We felt this was a reasonable value, since virtually all of the drugs in our sample had patent lifetimes significantly less In our sample, there is also a high overlap between the biopharmaceutical and orphan drug than 20 years, and products with substantial market sales are expected to face strong generic sets. This phenomenon has been discussed elsewhere and is the result of several factors.[11] First, competition and sales losses after patent expiration. While some products may have positive sales many of the initial biotech drugs were recombinant versions of natural hormones with approved after year 20, these sales are expected to be small in size and also would have very low weights in indications for small patient populations. In addition, many biopharmaceutical firms sought the any type of discounted present value analysis.
market exclusivity protection of orphan drug status given the initial uncertainties surrounding We used a two-step procedure to project future sales values for the products in this sample. A key time point in the sales life cycle is year of patent expiration. This is clustered in years 10 It is important to point out that there is a wide variance in the sales revenues realized by through 14 for the current sample. The first step in our approach involves projection to the year of orphan drugs in our sample. In particular, some of the novel biotech drugs granted orphan drug patent expiration, and the second step involves projection of the post-patent expiration values.
status were able to achieve blockbuster status by obtaining relatively large reimbursements per drug To project sales to the point of patent expiration, we utilized a similar approach to our past treatment. In addition, some of these drugs received orphan drug status for some indications as well analyses.[1,2] In particular, we constructed a reference life cycle curve based on the sales of products as approval for other "non-orphan" indications. Conversely, many of the orphan drug approvals in introduced in the mid-1980s (i.e., the new drug cohort immediately preceding the current one). We the 1988-92 period were for very rare conditions and had sales which are very small by historical used this as the basic framework to project sales values for most of the NCEs. However, to take standards (i.e., sales of only a few million dollars). Hence, the group of orphan drug compounds is account of recent market and competitive developments affecting the demand for the leading compounds and therapeutic groups, we also utilized the sales forecasts from a group of security analysts to make adjustments when there was a significant deviation from the reference case.
The estimated sales for the period after patent expiration were based on an analysis of The non-orphan drugs exhibit the general characteristics observed in prior work--rapid generic competition in the mid-1990s.[12,13] Using this analysis, the average sales percentage growth after launch, maturation about 10 to 11 years into the life cycle and then rapid sales decline decline in the first two years after patent expiration for products with 50 million dollar sales or after patent expiration and generic entry. By contrast, the orphan compounds exhibit more more at the time of patent expiration were computed to be 43% and 42%, respectively.3 Thereafter moderate growth rates after launch, a much lower expected peak sales level, but also slower a 10% annual decline is utilized over the remaining years of market life. In our analysis of generic expected rates of sales declines in the later stages of the life cycle. The latter phenomenon is due to competition since the passage of the 1984 Waxman-Hatch Act, we have observed a strong trend longer average patent protection periods as well as less generic exposure for the orphan drug over time toward increased sales erosion after patent expiration.[13] Since most of the products in population, given their smaller average sales levels.
our sample will be experiencing patent expiration in the early part of the next decade, the rates of As discussed, orphan drugs face different economic incentives in terms of the R&D and sales erosion after patent expiration for these drugs will tend to be understated if current trends regulatory process compared with non-orphan drugs. In future work, we plan to investigate their economic returns in a separate study. Nevertheless, the 28 orphan drugs in our current sample are very heterogenous and include some of the leading biopharmaceutical products such as EPO and IV. Empirical Results
human growth hormone. Because of this, we have chosen to retain these orphan drugs in our A. Sales of Orphan Versus Non-Orphan Drugs
sample but also to analyze the distribution of sales with and without these drugs present. The The first issue that we examine is the sales performance of the orphan versus non-orphan results do not change in a qualitative manner.
drugs in our sample. Figure 2 shows a plot of the life cycle sales profiles for the mean compound in these two sub-samples of drugs. As discussed in the previous section, the values in the first half of B. The Distribution of Sales for 1988-92 Introductions
the life cycle are essentially based on realized sales, while those in the second half are projected Figure 3 provides a plot of the expected sales profiles for the full sample of 1988-92 new using information on patent expiration and other inputs.4 drugs. This is an exact counterpart to Figure 1 and shows life cycle sales patterns for the top two deciles, the median and mean drug compounds. The main observed difference between Figures 1 3The probability of generic competition is low for drugs with sales at the time of patentexpiration that are below 50 million dollars. Accordingly, we assumed no generic competition will and 3 is associated with the top decile of drugs. In particular, the "mountain" type profile of the top occur in the case of these smaller-selling drugs.
decile in Figure 3 has grown taller and steeper compared to the other profiles displayed in these 4Since our sample involves a basket of new drugs introduced between 1988 and 1992, years 8through 11 of the life cycle are a blend of actual and forecasted sales. For example, year 8 involves the first year of forecasted sales for the 1992 cohort and actual sales for the 1988 through 1991 cohorts. Similarly, year 11 involves actual sales for the 1988 cohort and projected sales for the figures. In this respect, the top decile in the later time period exhibits more rapid rates of growth The top decile in Figures 3 and 4 is dominated by new drug introductions which are after launch, a peak that is more than 50 percent greater in real terms than for the 80-84 cohort and a pioneers or early entrants in a new therapeutic class of compounds. In particular, this group faster rate of expected sales decline after patent expiration.
includes the world's largest selling drug in 1998, Prilosec, the first drug in the proton pump The definite impression from Figure 3 is that the distribution of revenue has become more inhibitor class which is used to treat ulcers. It also includes the first two selective serotonin re- skewed over time. Of course, one factor contributing to this trend is the orphan drug phenomenon. uptake inhibitors, Prozac and Zoloft, used to treat depression. Also in the top decile of drugs are Most of the orphan drugs are concentrated at the bottom of the distribution, with a few blockbuster the two largest selling biopharmaceutical therapies--Epogen for treating anemia and Neupogen, drugs in the top decile. This tends to make the overall distribution more skewed. However, the which is used as an adjunct chemotherapeutic agent. In addition, the top-selling decile includes basic findings are not altered in a qualitative manner when one omits the orphan drugs from the Taxol, the leading chemotherapeutic drug for ovarian cancer; Norvasc, a new kind of calcium channel antagonist for treating hypertension; Biaxin and Zitromax, the first two introductions in the The movement toward more skewness over time, indicated in Figure 3, was confirmed by a Macrotide class of anti-infective agents, and Pravachol and Zocor, two leading statin drugs for more detailed analysis that we performed. This is presented in Figure 4. In this figure, we plot the full distribution of sales by decile for the 1980-84 and 1988-92 cohorts. Here the sales data are based on the seventh year after launch, so that this analysis is based completely on actual sales C. Changes in Mean Sales over Time
values. In particular, the top decile of 1988-92 new drugs account for 56 percent of overall sales In the case of skewed distributions, the revenue performance of the mean compound is revenue for the full sample of 110 drugs. If we omit the 28 orphan drugs, the top decile accounts disproportionately affected by the realized values in the upper tail of the distribution. Accordingly, for 52 percent of sales revenues. By contrast, the top decile of 1980-84 NCEs accounted for 48 we would expect the mean sales to be significantly greater in the 1988-92 cohort, compared to the percent of overall sales (and the same percentage of quasi-rents).
earlier 1980-84 one. In order to see how sales of the mean compound have changed over time, we plot the mean curves for the two time cohorts on a separate graph in Figure 5. This graph is based on the entire sample of 1988-92 drugs, including the orphan compounds. The case with the orphan drugs excluded is shown in Figure 6.
In both cases, there is a significant upward shift in the mean sales curves through the period of product maturity. However, the faster rate of generic competition expected for the later time cohorts causes a projected convergence of the two curves after year 15 of the life cycle. is especially the case for the group of large firms with R&D outlays between 300 and 500 million Nevertheless, the expected present value of sales revenues will be higher for the more recent time dollars. The total seventh year sales for the new drugs portfolio of these nine firms range from less cohort, given the positive differences in the earlier years of the life cycle. As noted, this is driven in than 100 million dollars to over 3 billion dollars.5 large part by the sales performance of the top decile products.
In prior work, we examined the temporal pattern of new drug sales outcomes across firms.[14] One of the main findings was that there was a high degree of variability from one period D. Sales of New Drug Introductions Versus R&D Outlays by Company
to the next in the performance in new drug sales of individual firms. This result is also consistent As discussed earlier, one important consequence of a skewed distribution is that even firms with the skewed distribution that we have observed for new drug introduction samples dating back with sizable portfolios of R&D projects can expect considerable variability in portfolio outcomes. In order to gain some further insights into this issue, we aggregated each company's sales (in the seventh year of market life) for all its new drug introductions in the 1988 to 1992 period. We then E. Comparisons with Global Introductions Case
plotted these portfolio outcomes against the company's pharmaceutical R&D expenditures in the In a recent analysis, William Machtiger of IMS International has examined the worldwide 1983-85 period. We utilized an average lag of six years between R&D expenditures and new drug sales levels in 1997 of 315 drugs that had their first introduction into the worldwide market over the introductions to reflect the long gestation period in pharmaceutical R&D.[8] period 1986-92.[15] This provides an interesting benchmark to compare with our sample of U.S.
Figure 7 shows the resulting plot of new drug sales versus R&D expenditures for a total of drugs. One would expect a global sample of NCE introductions to exhibit more skewness than a 18 firms for which R&D expenditure data were available. The firms span a considerable size U.S. based because many of these global introductions are "localized" NCEs. In particular, many of spectrum, but all of them can be characterized as multinational companies that are also vertically these NCEs are introduced in only a few countries with much smaller markets than the United integrated across all types of pharmaceutical activities (i.e., R&D, manufacturing and marketing). States. By contrast, U.S. NCE introductions tend to include a greater percentage of international or The annual R&D expenditures of these 18 firms in the mid-1980s was between 100 million and 500 million dollars (measured in 1992 $).
5We did not include the biotech group of firms in Figure 7 because annual R&D expenditures Figure 7 shows that there is a positive relationship between a company's R&D expenditures were not available for these firms for the 1983-85 period. However, we do know that these firmswere in their formative period and had annual R&D expenditures below 100 million dollars. and its subsequent sales from new drug introductions. However, there is also a lot of variation in Inclusion of these biotech firms would have produced much more scatter than shown in Figure 7. This is true because most of the biotech firms had very small sales from new drug introductions, but the scatter of points around the best fitted least squares regression line, as shown in Figure 7. This one firm in particular (Amgen) had annual sales in excess of 1.5 billion dollars in the seventh yearof market life from its two blockbuster products introduced in the 1988 to 1992 period.
In fact, the global sample does exhibit more skewness. In this regard, the drugs were performance of their new drug introductions. In this regard, we found that firms which spent grouped by their worldwide sales levels. The top group were the new drug introductions which between 300 and 500 million dollars for R&D in the mid-1980s (measured in 1992 dollars) had achieved worldwide sales of 500 million in 1997. Approximately 10 percent of the 315 global aggregate sales from their new product introduction in the 1988 to 1992 period ranging between introductions were in this category. This top decile accounted for 66 percent of the 60 billion 100 million and 3 billion dollars after 7 years of market life. Moreover, as one would expect in this dollars in total sales achieved by this group of 315 drug introductions. At the other end of the situation, there is a high degree of volatility across companies in the historical sales performance of distribution were products that achieved less than 50 million dollars in worldwide sales. These new drug cohorts from different time periods.
products accounted for 57 percent of the introductions but only 4 percent of the total sales. This is a Because pharmaceutical firms and venture capital firms confront outcome distributions with significantly greater degree of skewness than what was observed for our sample of U.S.
similar properties, they face similar strategic issues in managing the extreme risks and rewards that characterize their investment activities. In particular, timing and adaptability are critical elements for both entities. On the one hand, it is important to recognize and accelerate the development of V. Discussion and Conclusions
potential winners to get them to the market as quickly as possible. At the same time, it is important The analysis presented in this paper indicates that the distribution of sales revenues for new to terminate and cut one's losses on unsuccessful projects before large development costs become drug compounds is highly skewed in nature. In this regard, the top decile of new drugs accounts for totally sunk costs. This is a difficult balancing act that is subject to alternative types of errors. more than half of the total sales generated by the 1988 to 1992 cohort analyzed in this paper. Venture capital firms employ external financial controls including staged finance and milestones as Furthermore, the distribution of sales revenues for this cohort is more skewed compared with the risk-hedging strategies in this regard.[17,18] Pharmaceutical firms employ internal controls that 1980 to 1984 cohort that we analyzed in prior research.[1] This increased skewness can be embody some of these same principles.
explained in part by the emergence over the past decade and a half of orphan drug compounds In recent years, pharmaceutical firms have placed increased emphasis on reducing the under the incentives provided by the 1983 Orphan Drug Act. However, even when the orphan drug development time to market, using various strategies including parallel path development of compounds are omitted from the current sample, there still is some observed tendency toward activities that are along the critical path. In addition, firms are incorporating competitive and pharmacoeconomic analysis earlier in the development life cycle, well prior to the go-no-go One important implication of this distribution observed in pharmaceuticals is that even very decision for phase III development. In this regard, many drug firms traditionally placed primary large firms with sizeable portfolios of R&D projects are subject to substantial volatility in sales control of their R&D decisions in the hands of scientific and technical personnel until clinical tests were nearly completed. However, DiMasi recently has shown that the main reason that new drugs References
are abandoned in phase III is not on safety or efficacy grounds, but for economic reasons.[19] Hence, Grabowski H, Vernon J. "Returns to R&D on new drug introductions in the 1980s," J.
prioritization of R&D projects, using economic modeling as well as technical milestones, can increase the probability of economic success and reduce the likelihood of expensive late-stage Grabowski H, Vernon J. "A new look at the returns and risks to pharmaceutical R&D," failures.[20] Many firms appear to be moving in this direction.
From a public decision-making perspective, the winners among new drug introductions Scherer FM, Harkoff D, Kudies J. "Uncertainty and the size distribution of rewards from with blockbuster sales make a tempting target for price controls and profit constraints. However, as technological innovation," forthcoming in the Journal of Evolutionary Economics, 1999.
we have shown in prior analyses, expected returns are highly sensitive to the performance of these Scherer FM. New perspectives on economic growth and technological innovation, top decile drugs.[21] Imposing controls on today's drug winners will have adverse effects for future Washington, DC: Brookings Institution Press, Chapter 5, 53-88.
R&D investments, especially for the very risky projects that involve the most novel approaches to Grabowski H, Vernon J. "The determinants of pharmaceutical research and development expenditures," forthcoming in the Journal of Evolutionary Economics, 1999.
Bond R, Lean D. "Sales, promotion and product differentiation in two prescription drug markets," Washington, DC: Federal Trade Commission Staff Report, Government Printing Berndt E, Bui L, Lucking-Reily D, et al. "The roles of marketing, product quality and price competition in the growth and composition of the U.S. anti ulcer drug industry," in Bresnahan T and Gordon RJ, eds., The economics of new goods, Chicago, Ill: University of DiMasi J, Hansen R, Grabowski H, et al. "The cost of innovation in the pharmaceutical industry," J Health Econ 1991; 10: 107-42.
Grabowski H, Mullins CD. "Pharmacy benefit management, cost-effectiveness analysis and drug formulary decisions," Soc Sci Med 1997; 45(4): 535-54.
Sheila Shulman, Elaine M. Healy, Louis Lasagna, eds., PBMs: reshaping the Grabowski H. "The effect of pharmacoeconomics on company research and development pharmaceutical distribution network, New York: Haworth Press, 1998.
decisions," Pharmacoeconomics, May 1997, 11(5): 389-97.
Schulman S, Bienz-Tadmor B, Seo PS, et al., "Implementation of the orphan act: 1983-91," Grabowski H, Vernon J. Prospects for returns to pharmaceutical R&D under health care Food and Law J. 1992; 47(4): 363-404.
reform, in Helms R, ed., Competitive strategies in the pharmaceutical industry, Washington Grabowski H, Vernon J. "Longer patents for increased generic competition in the U.S.: the Waxman-Hatch Act after one decade," Pharmacoeconomics 1996; 10(supp 2): 110-123.
Grabowski H, Vernon J. "Effective patent life in pharmaceuticals," forthcoming in the International Journal of Technology Management, 1999.
Grabowski, H, Vernon J. "Innovation and structural change in pharmaceuticals and biotechnology," Industrial and Corporate Changes 1994; 3(2): 435-449.
Machtiger WS. "The strategic management review of the world pharmaceutical market presentation to Drug Information Association Workshop, London, October 13, 1998.
Thomas LG III. Industrial policy and international competitiveness in the pharmaceutical industry, in Helms R. ed., Competitive strategies in the pharmaceutical industry, Washington DC: AEI Press, 1996: 107-129.
Sahlman W. "The structure and governance of venture capital organizations," Journal of Simpson H. Biotechnology and the economics of discovery in the pharmaceutical industry, London: Office of Health Economics, 1998, ch. 3.
DiMasi JA. "Success rates for new drugs entering clinical testing in the United States," Clin Venture Economics(383 Start-up investments) 1st Decile
2nd Decile
ons
illi 200
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
Sales Year
9
1
of 100
ons
illi
9 10 11 12 13 14 15 16 17 18 19 20
Non-Orphan NCEs
Orphan NCEs
1st Decile
2nd Decile
illions
M
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
Sales Year
80-84 NCEs
88-92 NCEs
1992 $ 100
illions
M
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
1988-1992 NCEs
1980-1984 NCES
9
1
of 100
ons
illi
9 10 11 12 13 14 15 16 17 18 19 20
1988-1992 NCEs
1980-1984 NCES
Sales for 1988-92 New Drug Introductions
Plotted against 1983-87 R&D Expenditures by Company
(in millions of 1992 dollars)
g
ru
D
w 3000
les f
a
S
y 1000
Company R&D Outlays (1983-85 Average)

Source: http://www.cric.ac.uk/cric/events/schumpeter/papers/54.pdf